1. Ron Trompert – Testbed1 Software – 7 November 2001 - n° 1 Partner Logo Testbed1 Software Ron Trompert ron @ sara.nl

2. Ron Trompert – Testbed1 Software- 7 November 2001 - n° 2 Summary u Testbed1 software presentation at CERN n Testbed overview n Job submission n User interface n Information system n WP4 stuff n EDG Globus Configuration u Globus2.0 installation on a Compaq cluster u SARA’s contribution to testbed1

3. Ron Trompert – Testbed1 Software- 7 November 2001 - n° 3 Testbed overview UI JDL L&B Job Submit Event RB Output Sandbox Input Sandbox JSS SE CE Brokerinfo Output Sandbox Input Sandbox II Job Status WP2/ReplicaCatalogue

4. Ron Trompert – Testbed1 Software- 7 November 2001 - n° 4 Job Submission # # ----- Job Description File ----- # Executable = "WP1testC"; StdInput = "sim.dat"; StdOutput = "sim.out"; StdError = "sim.err"; InputSandbox = {"/home/wp1/HandsOn-0409/WP1testC","/home/wp1/HandsOn-0409/file*”, "/home/wp1/DATA/*"}; OutputSandbox = {"sim.err","test.out","sim.out"}; Rank = other.AverageSI00; Requirements = (other.OpSys == "Linux RH 6.1" || other.OpSys == "Linux RH 6.2”) && (other.RunTimeEnvironmnet == “CMS3.2”); InputData = "LF:test10096-0009"; ReplicaCatalog = "ldap://sunlab2g.cnaf.infn.it:2010/rc=WP2 INFN Test Replica Catalog,dc=sunlab2g, dc=cnaf, dc=infn, dc=it"; DataAccessProtocol = "gridftp";

5. Ron Trompert – Testbed1 Software- 7 November 2001 - n° 5 User interface u dg-job-submit allows the user to submit a job for the execution on remote resources n -r, -resource res_id the job is submitted by the Broker to the resource identified by res_id n -i, -input input_file the user must choose a resource id from a list of resources n -n, -notify e_mail_address an e-mail message is sent to the specified e-mail address when the job enters in one of the following status: READY, RUNNING, ABORTED or DONE n -o, -output out_file the generated dg_jobid is written in the file out_file

6. Ron Trompert – Testbed1 Software- 7 November 2001 - n° 6 u dg-job-get-output requests to the Broker the job output files, specified by the OuputSandbox attribute of the job-ad, and stores them on the submitting machine local disk u dg-job-list-match returns the list of resources which fulfills job requirements u dg-job-cancel cancels one or more submitted jobs u dg-job-status displays bookkeeping information about submitted jobs u dg-job-get-logging-info displays logging information about submitted jobs

7. Ron Trompert – Testbed1 Software- 7 November 2001 - n° 7 Information system u There are two to pick from n Globus MDS 2.1 s Integrated in Globus s Memory caching, LDAPv3, GSI authentication s Multiple VO’s on one node s GRIS and GIIS use the same slapd and listen to the same port. They are only distinguished through their DN’s: n GIIS: … Mds-Vo-Name=VO_NAME, o=Grid n GRIS: … Mds-Vo-Name=local, o=Grid n OpenLDAP Ftree s Runs independently of globus s Memory caching, LDAPv3, GSI authentication s GRIS and GIIS on different ports

8. Ron Trompert – Testbed1 Software- 7 November 2001 - n° 8 WP4 stuff u Interim Installation System n Image Cloning s Quick and easy installation of identical machines in large clusters n Local ConFiGuration system s Handle automated installation and configuration in a very diverse and evolving environment s Cfengine-like

9. Ron Trompert – Testbed1 Software- 7 November 2001 - n° 9 Image cloning Image server Boot server Configuration Management System Reference node New node

10. Ron Trompert – Testbed1 Software- 7 November 2001 - n° 10 ldxprof Generic Component Generic Component rdxprof LCFG Components DBM File LCFG configuration files mkxprof Web Server XML Profile (one per client node) LCFG SERVER HTTP LCFG CLIENT LCFG

11. Ron Trompert – Testbed1 Software- 7 November 2001 - n° 11 LCFG: RPM configuration u The list of RPMs to be installed on a node is defined in text files stored in the LCFG server u These RPM lists can be shared by several nodes; the LCFG configuration associates one list to each node u These files must be accessible to the clients using a remote file system (NFS) u The RPMs themselves must be also accessible via NFS AUTOMATICALLY REMOVED u Any RPM installed manually in the client without being included in the files will be AUTOMATICALLY REMOVED

12. Ron Trompert – Testbed1 Software- 7 November 2001 - n° 12 EDG Globus configuration u Everything is installed in $GLOBUS_LOCATION i.e. /opt/globus u Security files reside in /etc/grid-security n CA certificates, CRLs and signing policy files in /etc/grid- security/certificates n Grid-mapfile, host key and host certificate in /etc/grid-security u Globus setup packages are not used. They are replaced by the EDG config packages globus_ -edgconfig where is the name of the globus setup package it replaces

13. Ron Trompert – Testbed1 Software- 7 November 2001 - n° 13 u A single configuration file /etc/globus.conf n Local (internal) configuration files are (re)created by each restart of service with parameters from /etc/globus.conf n Simple format: s # at start of line means a comment s MACRO=VALUE s Example: s GLOBUS_LOCATION=/opt/globus

14. Ron Trompert – Testbed1 Software- 7 November 2001 - n° 14 Globus2.0 installation on a Compaq cluster u Build_from_source_tarballs n Perl script to build the distribution n Site specific information should be set: s my $SRC_TAR_LOC = “ … “; s my $SRC_LOC = “ …”; s my $GLOBUS_LOC = “ … “; s my $FLAVOR = “gcc32dbg”; n Although the Compaq is a 64-bit machine 32-bit was specified here because when you use 64, the scripts think that you are on a SGI machine and invoke the MIPSpro --enable-64bit compiler option. 32 does nothing. n Haven’t tried vendorcc (ccc) yet s my $LOG_LOC = “ … “; s my $TMP_DIR = “ … “;

15. Ron Trompert – Testbed1 Software- 7 November 2001 - n° 15 u Compilation n Globus-ssl-utils-2.1 s Error message: undefined reference to ‘_OtsRemainder64Unsigned’ n Linking with –lots by editing the configure script s Warning: gnu_regex.c:cast from pointer to integer of different size n On the Compaq this did not seem to cause any problems. The machine only has 512Mb of memory and tests showed that the high 32 bits of 64-bit pointers did not change. n Globus_gatekeeper.c & globus_gram_job_manager.c s Error: redefininition of sys_errlist[ ] n For some reason “TARGET_ARCH_LINUX” was not #define-ed. A dirty hack solved the problem. u Globus-script-pbs-submit n #PBS –l ncpus=$grami_count should be #PBS –l nodes=$grami_count n Set is_cluster=true

16. Ron Trompert – Testbed1 Software- 7 November 2001 - n° 16 u Tell Globus where MPI is in globus-job-manager-tools.sh and globus- sh-tools.sh u LD_LIBRARY_PATH containing $GLOBUS_LOCATION/lib for globus-gatekeeper and in.ftpd n Solved by wrapper

17. Ron Trompert – Testbed1 Software- 7 November 2001 - n° 17 SARA’s contribution to testbed1 u Resources n Compute Elements s Compact n 16 nodes 600MHz alpha n Myrinet 2000 interconnect n Fast Ethernet interconnect n OpenPBS n GridFTP n Jobmanagers fork and pbs n MPI over myrinet and fast Ethernet. MPI over myrinet is the default, MPI over Fast Ethernet only through RSL. s Itanium cluster?

18. Ron Trompert – Testbed1 Software- 7 November 2001 - n° 18 SARA’s contribution to the testbed1 n Storage Element s Teras with StorageTek NearLine storage n 10TB disk n 120 TB tape n GridFTP n No jobmanagers